Need a book? Engineering books recommendations...

Return to index: [Subject] [Thread] [Date] [Author]

Probability

[Subject Prev][Subject Next][Thread Prev][Thread Next]
Does anyone have an easy (key word) explanation as to why we have such 
radically different probabilities for the design earthquake event as opposed 
to the design wind or snow event.

To keep it simple just compare seismic to wind, and ignore importance 
factors.  Both induce lateral loads that have been traditionally treated the 
same as far as load factors.  Currently we design for a seismic event with a 
10% chance of exceedance in 50 years and we will soon make the design 
earthquake a 2% chance of exceedance in 50 years (inside sources).

On the other hand, the chance of exceedance with wind (Re: ASCE 7-95) is 64% 
in 50 years.  I understand that the intended performance under the 
respective design event is linear with wind and nonlinear with seismic.  But 
how do you explain the design case probabilities to an owner if you are 
evaluating an important facility that is both highly seismic and has high 
winds.  Examples of such areas are Taiwan, Guam, and Puerto Rico.

In consulting with wind specialists, the generally recommended exceedance 
probability drops depending on which wind engineer you talk to and what the 
facility is.  The Federal Energy Regulator Commission lowers the probability 
for wind velocities for liquefied natural gas facilities to 5% in 50 years. 
 (This value is implied by the 10,000 year recurrence interval.)

Wind engineers have advised me in the past to drop the exceedance to 40% for 
school emergency shelters and critical water supply facilities.  This is 
still a long way off from 2% in 50 years.

When evaluating high risk structures (nuclear, liquefied natural gas, etc.) 
the probabilities for natural hazards are very similar.  For buildings and 
building codes there is much greater divergence.  Why?

I would appreciate any insight.

Harold Sprague, spragueho(--nospam--at)bv.com
Black & Veatch